The full text of this item is not available at this time because the student has placed this item under an embargo for a period of time. The Libraries are not authorized to provide a copy of this work during the embargo period, even for Texas A&M users with NetID.
A Cognitive Assistant for Human-AI Collaborative Early Mission Formulation
Abstract
The popularity of Cognitive Assistants, defined as intelligent AI agents that enhance human capabilities and interact through natural language, has surged in the last decade. Recent advances in natural language understanding have allowed the technology to showcase its real potential.
Cognitive Assistants are well suited to solving the issue of cognitive overload, where a human has too much information and tasks to process, leading to mistakes or suboptimal decisions. This issue is faced in the workplace by many workers, including aerospace systems engineers. Concur-rent design sessions and large trade studies performed during the early design of complex systems such as space missions are tasks that may result in cognitive overload.
This work presents a new approach to address the issue of cognitive overload in systems engineers by creating a cognitive assistant catered to the use case of early mission design tasks. Daphne specializes in collaborating with engineers during the tradespace exploration process, an early design task about exploring a large set of possible designs for a system to choose the best options to study in more detail. Daphne interacts with the engineer through roles, where each role implements functions and expertise a systems engineer would seek from a human peer.
This work also describes a theoretical framework that models the collaboration between a human engineer and a cognitive assistant during the tradespace exploration process, modeling both as intelligent agents. The premise of the framework is that for cognitive assistants to be effective, they must understand the characteristics, state, goals, and knowledge of the human designer and adapt to them.
Finally, in order to validate the approach and the framework, this work details the results of four human subjects experiments where Daphne and the predictions from the framework are tested with novice and expert systems engineers. The first experiment suggests that using Daphne leads to increased task performance, while the second shows that no Daphne role has a significant effect on performance compared to others. The last two experiments show that adapting to the goals and the expertise of the human engineer can lead to better performance metrics and learning.
Subject
Artificial IntelligenceEngineering Design
Systems Engineering
Natural Language Programming
Human Subjects Studies
Space Systems
Earth Observation
Remote Sensing
Virtual Assistants
Cognitive Assistants
Citation
Viros i Martin, Antoni (2022). A Cognitive Assistant for Human-AI Collaborative Early Mission Formulation. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /197919.